Aerial Imagery Pixel-level Segmentation

12/03/2020
by   Michael R. Heffels, et al.
0

Aerial imagery can be used for important work on a global scale. Nevertheless, the analysis of this data using neural network architectures lags behind the current state-of-the-art on popular datasets such as PASCAL VOC, CityScapes and Camvid. In this paper we bridge the performance-gap between these popular datasets and aerial imagery data. Little work is done on aerial imagery with state-of-the-art neural network architectures in a multi-class setting. Our experiments concerning data augmentation, normalisation, image size and loss functions give insight into a high performance setup for aerial imagery segmentation datasets. Our work, using the state-of-the-art DeepLabv3+ Xception65 architecture, achieves a mean IOU of 70 validation set. With this result, we clearly outperform the current publicly available state-of-the-art validation set mIOU (65 Furthermore, to our knowledge, there is no mIOU benchmark for the test set. Hence, we also propose a new benchmark on the DroneDeploy test set using the best performing DeepLabv3+ Xception65 architecture, with a mIOU score of 52.5

READ FULL TEXT

page 3

page 4

page 6

page 16

page 20

page 21

page 22

page 23

research
09/25/2017

Fast Vehicle Detection in Aerial Imagery

In recent years, several real-time or near real-time object detectors ha...
research
06/01/2023

Geo-Tiles for Semantic Segmentation of Earth Observation Imagery

To cope with the high requirements during the computation of semantic se...
research
12/21/2021

Mapping industrial poultry operations at scale with deep learning and aerial imagery

Concentrated Animal Feeding Operations (CAFOs) pose serious risks to air...
research
04/24/2017

Using Global Constraints and Reranking to Improve Cognates Detection

Global constraints and reranking have not been used in cognates detectio...
research
10/28/2019

A data set of aerial imagery from robotics simulator for map-based localization systems benchmark

Purpose: This paper presents a new dataset of Aerial Imagery from Roboti...
research
11/27/2022

Performance evaluation of deep segmentation models on Landsat-8 imagery

Contrails, short for condensation trails, are line-shaped ice clouds pro...

Please sign up or login with your details

Forgot password? Click here to reset